| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
Georgetown Center for Food and Nutrition Policy, Georgetown University, Washington, DC
Address reprint requests to: Maureen L. Storey, PhD, Georgetown Center for Food and Nutrition Policy, 3240 Prospect Street, NW, Washington, DC 20007. E-mail: storeym{at}georgetown.edu.
 |
ABSTRACT |
|---|
 TOPFOOTNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSCONCLUSIONSACKNOWLEDGMENTSREFERENCES |
|---|
Methods: Data from the USDA 1994–96 Continuing Survey of Food Intakes by Individuals were used in multivariate regression analyses to assess the statistical and practical significance of added sugars intake for diet and nutrient adequacy.
Results: The association of added sugars with consumption of vitamins, minerals and servings of foods in the USDA Food Guide Pyramid was usually statistically significant. For the model of all individuals over two years of age, individuals who consume more added sugars are predicted to consume more grains, lean meat and iron and to consume fewer vegetables and fruits and less dairy, vitamin A, calcium and folates. Children who consume more added sugars are predicted to consume more grains, vitamin C, iron and folates and to consume less dairy. Adolescents who consume more added sugars are predicted to consume more grains, vitamin C and iron and less fruit.
Conclusion: The associations, whether positive or negative, however, were always small from either a practical perspective or in comparison to the associations of other sources of energy.
Key words: nutrient displacement, sugars, dietary guidelines, food guide pyramid
|
INTRODUCTION |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSCONCLUSIONSACKNOWLEDGMENTSREFERENCES |
|---|
In spite of these reports, there is contention that foods and beverages with added sugars dilute micronutrients and displace nutrient-dense foods in the diet. This emotional debate about the role of added sugars in diet quality regrettably lends credibility to the spurious distinction that the mono- and disaccharides in processed foods are absorbed, metabolized and utilized differently than those occurring naturally in grains, fruits, vegetables and dairy foods.
The controversy may have been fueled partly by the method used by the U.S. Department of Agriculture (USDA) to quantify the intake of foods and some components of foods, as illustrated by the USDA’s Food Guide Pyramid (FGP). For example, the sucrose added to yeast breads is placed in the tip of the FGP; on the other hand, the sucrose found in fruit is not. Thus the FGP leaves the impression that the body distinguishes between the sucrose from the two different sources. In addition, the USDA’s method of calculating added sugars inflates the amount of sucrose in breads because there is little sucrose left in the final food due to the fermentation by the yeast and the formation of other compounds (e.g., Maillard compounds) during baking.
The role of added sugars in health was contested during the deliberations of the 2000 U.S. Dietary Guidelines Advisory Committee (DGAC). Members of the committee argued that government guidelines should encourage consumers to reduce their consumption of sugars, particularly added sugars, in an effort to reduce escalating obesity rates [8]. The committee claimed that consumption of added sugars may lead to obesity and may displace other more nutrient-dense foods from the diet. The DGAC cited a few recent studies in support of the nutrient displacement claim [9,10,11,12], but they also cited other studies [13,14] that showed "no consistent associations between intake of total sugars and nutrient adequacy."
The DGAC heard conflicting testimony that consumption of added sugars has increased alarmingly at the same time that obesity rates have increased, but the committee also heard testimony that it has not. As pointed out previously, consensus reports conclude there is no association between sugars intake and obesity. Moreover, Gibson [15] stated: "There is little evidence that diets high in total sugars are associated with obesity."
The DGAC and its chair, Cutberto Garza, struggled with the mixed and limited evidence on the effect of added sugars on obesity and diet quality. "I think this guideline has proven the most difficult and troublesome to both departments for us because in fact the data that we’ve had to work with has really been appalling in many respects," reflected Chairman Garza near the end of the deliberations on the sugars guideline [16].
Despite the limitations of the evidence, the DGAC recommended several changes in the guideline for sugars. First, the committee recommended changing the 1995 guideline that read "choose a diet moderate in sugars." Second, the DGAC recommended that the sugars guideline distinguish added sugars from naturally occurring sugars [17,18].
The purpose of our study was to examine the association between added sugars intake and consumption of vitamins, minerals and servings of foods in the USDA Food Guide Pyramid. We developed a behavioral model of the food choices people make and the effect that those choices have on diet quality. One important feature of our model is that we include statistical controls for all of the possible sources of energy for an individual. This provides a more complete model specification and allows comparisons of the relative effects of added sugars, carbohydrates minus added sugars, fat, protein and alcohol on diet quality.
|
MATERIALS AND METHODS |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSCONCLUSIONSACKNOWLEDGMENTSREFERENCES |
|---|
Data Analysis
A series of multivariate linear regression models was developed to predict: (1) consumption of servings of the major food groups in the USDA FGP and (2) the percentage of the RDA of selected vitamins and minerals consumed by all individuals over age two, children ages 6 to 11 years and adolescents ages 12 to 19 years. The models are based on the grams of added sugars, carbohydrates minus added sugars, protein, fat and alcohol consumed by respondents. The models also controlled for age and gender and were estimated using robust standard errors that accounted for the complex survey design of CSFII.
The equation for our model is:
 |
 |
 |
 |
 |
 |
 |
 |
 |
The independent variables were chosen to represent all of the sources of an individual’s total energy intake. It is important to include all of the diet variables in order to avoid specification error. We know that the components of an individual’s diet are systematically related to one another; for example, the percent of energy from fat and the percent of energy from carbohydrates typically have an inverse relationship. In addition, each of these diet variables is likely to be related to our dependent variables (the number of servings from the major food groups and the percentage RDA of calcium, vitamin A, vitamin C, iron and folates consumed). Since these diet variables are correlated with one another and with our dependent variables, they were included in our regression model. Otherwise, the uncontrolled effects of the excluded variables would bias the estimated effects of the diet variables included in the model. Details on the problems of specification error may be found in Gujarati [20] or other intermediate to advanced epidemiology or biostatistics textbooks.
Only about one percent of the children in our sample reported consumption of one or two grams of alcohol. We included alcohol in our models for children because alcohol is an energy source and excluding it would introduce bias in the coefficient estimates of the other independent variables. The coefficient estimates for alcohol consumption in the model for children should be interpreted with caution, given the small number of children who reported alcohol consumption.
Future work will expand the set of control variables used in these models. For example, some recent studies have identified ethnic differences in food consumption patterns [21,22], and this will be a useful distinction in future research projects.
In the models of consumption of the major food groups, the coefficients for these energy sources, bj, represent the predicted change in the number of servings of a major food when the reported consumption of the energy source increases by one (1) gram. For example, if the coefficient for added sugars, b1, was equal to 0.01 we could say that an additional gram of added sugars in the diet produces a 0.01 increase in the predicted number of servings for that food group. The coefficient for an individual’s age, b6, represents the predicted change in the number of servings for a one-year increase in age. The coefficient for the gender variable female, b7, represents the predicted difference in the number of servings consumed by females as compared to males.
The coefficients are interpreted similarly in the models of the percentage of the RDA of vitamins and minerals consumed by an individual. The coefficients for these energy sources, bj, represent the predicted change in the percentage of the RDA consumed when the reported consumption of the energy source increases by one gram. For example, if the coefficient for added sugars, b1, was equal to 0.1, we could say that an additional gram of added sugars in the diet produces a 0.1 increase in the predicted percentage of the RDA for that vitamin or mineral. The coefficient for an individual’s age, b6, represents the predicted change in the percentage of the RDA for that vitamin or mineral for a one-year increase in age. The coefficient for the gender variable female, b7, represents the predicted difference in the percentage of the RDA for that vitamin or mineral consumed by females as compared to males.
|
RESULTS |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSCONCLUSIONSACKNOWLEDGMENTSREFERENCES |
|---|
The association of added sugars with the consumption of the major food groups and the vitamins and minerals examined was inconsistent and varied with age group (Fig. 1a–f). The results ranged from no association to a statistically significant association that was positive or negative. Even the statistically significant effects were so small that they are likely to be of no clinical importance.
|
Protein and fat had mixed associations with the consumption of the major food groups, vitamins and minerals examined in this paper. The associations were usually statistically significant, but the size and direction varied depending on which food group, vitamin or mineral was examined.
Alcohol had a consistently negative, statistically significant association with consumption of the major food groups, vitamins and minerals examined in this paper. Compared to the other variables in the model, alcohol had a strong negative association with the variables examined. Moderate alcohol consumption was unlikely to have any serious effect on an individual’s diet, but heavy alcohol consumption tended to be associated with a poor diet.
The associations of added sugars with the examined outcome variables are shown in Fig. 1a–f. Most, but not all, of these results are statistically significant at less than the 0.05 level. Subsequent sections covering specific food groups and Tables 1a, 1b, 2a, 2b, 3a and 3b contain the complete reports on statistical significance. In general, additional grams of added sugars were associated with an increase in grain and lean meat consumption and an increase in the percentage RDA of vitamin C, iron and folates. Conversely, increased consumption of added sugars was generally associated with decreased servings of vegetables, fruit and dairy. The association of added sugars with the percentage RDA of vitamin A and calcium varied depending on the age group. Added sugars had a negative association with vitamin A consumption for all ages and children, but a positive association for adolescents. Added sugars were not associated with calcium consumption for children or adolescents, but had a slight negative association for all ages.
|
|
|
|
|
|
The following subsections discuss our findings for each major food group, vitamin and mineral analyzed.
Grains
Added sugars in the diet were associated with increased servings of grains. An extra gram of added sugars increased the predicted number of grain servings by 0.0052 (p<0.01) for all individuals, by 0.0043 (p<0.01) for children and by 0.0036 (p<0.01) for adolescents.
As one would expect, carbohydrates minus added sugars had the largest impact on an individual’s consumption of grain servings. An extra gram of carbohydrates increased the predicted number of grain servings by 0.031 (p<0.01) for all individuals, by 0.032 (p<0.01) for children and by 0.033 (p<0.01) for adolescents.
Fat also had a positive, statistically significant association with the predicted number of grain servings. Protein was negatively associated with grain servings for all individuals and adolescents, but not for children. Alcohol had the largest negative association. An additional gram of alcohol reduced the predicted number of grain servings by 0.0146 (p<0.01) for all individuals and by 0.0103 (p<0.01) for adolescents. The predicted association of alcohol with the consumption of grains by children was not statistically significant.
Vegetables
Added sugars tended to be associated with lower consumption of vegetables. Holding other variables constant, our model predicted that a one gram increase of added sugars consumption was linked to a decrease in the predicted number of vegetable servings by 0.0014 (p<0.01) for all individuals. The coefficients for children and adolescents were not statistically significant.
Grams of fat had the largest association with servings of vegetables. Each additional gram of fat was associated with an increase in the predicted number of servings of vegetables by 0.0120 (p<0.01) for all age groups examined. Grams of protein and carbohydrates minus added sugars also had positive and statistically significant associations with the number of servings of vegetables for all ages and for adolescents. Alcohol was not associated with vegetable servings.
Fruits
Consumption of added sugars had a weak, negative association with the predicted number of fruit servings for adolescents and all ages. An extra gram of added sugars decreased the predicted number of fruit servings by 0.0021 (p<0.01) for all individuals and by 0.0012 (p<0.05) for adolescents. Added sugars intake had no association with fruit consumption among children.
Carbohydrates minus added sugars had the strongest association with the number of servings of fruit. Each additional gram of carbohydrates minus added sugars increased the predicted number of fruit servings by 0.0181 (p<0.01) for all individuals, by 0.0192 (p<0.01) for children and by 0.0133 (p<0.01) for adolescents. Protein, fat and alcohol were all negatively associated with fruit consumption (p<0.05). It should be noted that, with one exception, in which the coefficient for the association of alcohol with the fruit consumption of children is not statistically significant, the negative associations of protein, fat and alcohol consumption with servings of fruit are much greater than that of added sugars.
Dairy
Added sugars had no association with dairy consumption by adolescents. An extra gram of added sugars decreased the predicted number of dairy servings by 0.0006 (p<0.05) for all individuals and by 0.0017 (p<0.05) for children. Even though added sugars had a statistically significant relationship with dairy consumption, this association is unlikely to be of practical importance. It would require huge behavioral changes for there to be any clinical or practical significance. In order to reduce the servings of dairy by one, individuals would need to consume an additional 1,667 grams of added sugars or 417 teaspoons of table sugar. Children would have to consume an additional 588 grams of added sugars or 147 teaspoons of table sugar.
Protein had the largest positive association with dairy consumption. An additional gram of protein was associated with a 0.0088 (p<0.01) increase in the predicted number of servings of dairy for all individuals, a 0.0211 (p<0.05) increase for children and a 0.0094 (p<0.05) increase for adolescents. Carbohydrates less added sugars also had a positive association with dairy consumption for all individuals, children and adolescents. Fat intake was positively associated with dairy consumption among all individuals only.
Alcohol consumption had the strongest negative association with dairy consumption. An additional gram of alcohol reduced the predicted number of dairy servings by 0.0054 (p<0.01) for all individuals and by 0.0044 (p<0.05) for adolescents. The coefficient for children was not statistically significant. Using our model, the data show that an additional gram of alcohol was roughly nine times worse for dairy consumption than was an additional gram of added sugars.
Lean Meat
Added sugars in the diet were associated with increased servings of lean meat for all individuals over age two, but there was no statistically significant association with intake among children or adolescents. An extra gram of added sugars increased the predicted number of lean meat servings by 0.0007 (p<0.05) for all individuals. The coefficients for children and adolescents were not statistically significant.
Not surprisingly, protein had the strongest association with consumption of lean meat. An additional gram of protein increased the predicted number of servings of lean meat by 0.1022 (p<0.01) for all individuals, by 0.0884 (p<0.01) for children and by 0.1036 (p<0.01) for adolescents. Fat and carbohydrates minus added sugars were generally associated with fewer servings of lean meat. Alcohol consumption had a statistically significant, positive association with the predicted number of servings of lean meat for all individuals, but it was not statistically significant for children or adolescents.
Vitamin A
Only the model for all individuals showed a negative association between added sugars and the percentage RDA of vitamin A consumed. An extra gram of added sugars decreased the predicted percentage of RDA for vitamin A by 0.1992 (p<0.01) for all individuals. The coefficients for children and adolescents were not statistically significant.
Fat and alcohol had the largest negative links with vitamin A consumption. A one-gram increase in fat consumption reduced the predicted percentage RDA for vitamin A by 0.7610 (p<0.01) for all individuals, by 0.6713 (p<0.01) for children and by 1.2210 (p<0.01) for adolescents. An additional gram of alcohol reduced the predicted percentage RDA of vitamin A by 0.5564 (p<0.01) for all individuals and by 0.2332 (p<0.01) for adolescents. The coefficient for alcohol on children’s consumption of vitamin A was not statistically significant. Using the results for all individuals, a gram of fat had almost a four times greater association with vitamin A intake than did a gram of added sugars; a gram of alcohol had almost a three times greater association with this vitamin than a gram of added sugars did.
Carbohydrates minus added sugars and protein had positive associations with the percentage RDA of vitamin A. An additional gram of carbohydrates minus added sugars increased the predicted percentage RDA of vitamin A by 0.7667 (p<0.01) for all individuals, by 0.9322 (p<0.01) for children and by 0.4552 (p<0.01) for adolescents. An additional gram of protein increased the predicted percentage RDA of vitamin A by 0.6399 (p<0.01) for all individuals, by 0.6499 (p<0.05) for children and by 1.9328 (p<0.01) for adolescents.
Vitamin C
Added sugars had a small, positive association with vitamin C consumption. An extra gram of added sugars increased the predicted percentage of RDA for vitamin C by 0.4167 (p<0.01) for children and by 0.1796 (p<0.05) for adolescents. The coefficient for all individuals was not statistically significant.
Fat and alcohol had a negative association with the predicted consumption of vitamin C. An additional gram of fat reduced the percentage RDA of vitamin C by 0.9865 (p<0.01) for all individuals, by 1.3956 (p<0.01) for children and by 0.8121 (p<0.01) for adolescents. An additional gram of alcohol reduced the percentage RDA of vitamin C by 0.3736 (p<0.01) for all individuals, by 101.6215 (p<0.01) for children and by 0.5100 (p<0.01) for adolescents.
Carbohydrates minus added sugars had a strong, positive and statistically significant association with vitamin C consumption across all three models. An additional gram of carbohydrates minus added sugars increased the predicted RDA of vitamin C by 1.2199 (p<0.01) for all individuals, by 1.6190 (p<0.01) for children and by 1.0817 (p<0.01) for adolescents. Protein was not related to vitamin C consumption.
Calcium
The association of added sugars with calcium consumption depended on the age group we examined (Figures 2a–c). Added sugars had no association with calcium consumption among children or adolescents, but there was a slight negative association among all individuals over two years of age. In the model for all individuals, a one-gram increase in added sugars reduced the predicted percentage RDA of calcium by 0.0316 (p<0.01).
|
Protein, carbohydrates minus added sugars and fat were all positively associated with the percentage RDA of calcium consumed for all individuals and adolescents. Protein and carbohydrates minus added sugars were positively associated with the percentage RDA for calcium among children.
Iron
Added sugars had a small, positive association with iron consumption. An extra gram of added sugars increased the predicted percentage RDA of iron by 0.0493 (p<0.01) for all individuals, by 0.1196 (p<0.01) for children and by 0.0901 (p<0.01) for adolescents.
Fat and alcohol had moderate, negative associations with iron consumption. An additional gram of fat in the diet reduced the predicted RDA of iron by 0.2557 (p<0.01) for all individuals, by 0.3687 (p<0.01) for children and by 0.3878 (p<0.01) for adolescents. Increasing alcohol consumption by one gram reduced the predicted RDA of iron by 0.2384 (p<0.01) for all individuals and by 0.1276 (p<0.01) for adolescents. The coefficient for the association of alcohol with iron consumption for children was not statistically significant.
Protein and carbohydrates minus added sugars were positively associated with iron consumption.
Folate
The results for folate consumption depended on the age group we examined. For the sample of all individuals, those who consume more added sugars tended to consume a lower percentage RDA of folates. An extra gram of added sugars reduced the predicted RDA of folate by 0.2457 (p<0.01) for all individuals. An additional gram of added sugars increased the predicted percentage RDA of folates by 0.1961 (p<0.01) for children. The relationship between folates and added sugars was not statistically significant for the adolescents.
Fat had the strongest negative association with folate consumption. An additional gram of fat reduces the predicted percentage RDA of folates by 0.5697 (p<0.01) for all individuals, by 1.0018 (p<0.01) for children and by 0.5602 (p<0.01) for adolescents. Alcohol had a negative association with folate consumption in the model for all ages. For those individuals two years old or older, an additional gram of alcohol reduces the percentage RDA of folates by 0.3495 (p<0.01). The coefficients for the association of alcohol with folate consumption for children and adolescents were not statistically significant.
Carbohydrates minus added sugars increased the predicted percentage RDA of folates. An additional gram of carbohydrates minus added sugars increased the predicted percentage RDA of folates by 0.8713 (p<0.01) for all individuals, by 1.5723 (p<0.01) for children and by 0.7460 (p<0.01) for adolescents. Protein was not related to folate consumption for all individuals, but folate consumption was positively linked with protein for children and adolescents.
|
DISCUSSION |
|---|
In our models, age and gender have strong relationships with the consumption of food groups, vitamins and minerals. We found that as age increased, dairy, vitamin C, calcium and folate intake decreased in the models for children and adolescents. Age was also negatively associated with fruit, vitamin A, iron and folates in the model for children. Age was not associated with fruit, vitamin A or iron intake among adolescents. Age had a positive relationship with lean meat consumption among both children and adolescents, and age was positively associated with consumption of grains and vegetables among adolescents.
Gender was also strongly associated with diet quality. Adolescent girls consumed more fruit, vitamin A and vitamin C, but fewer grains and less iron than boys did. Girls ages 6 to 11 consumed less iron and folates than did boys of the same age. When all other variables were controlled, gender was not correlated with the other variables in these two models. Among all individuals, being female was associated with less grain, meat, calcium, iron and folates and more vegetable, fruit, vitamin A and vitamin C intake. Being female was not associated with consuming fewer servings of dairy foods.
Others have found similar results. In one study, Bowman [24] showed a statistically significant decrease in micronutrient intake among all individuals over the age of two consuming 18 percent or more of their energy from added sugars. Yet this study also showed that these individuals met or exceeded 100 percent of the 1989 RDA for vitamin A, vitamin C, thiamin, riboflavin, niacin, folates, vitamin B12, phosphorus and iron. Moreover, they exceeded 77 percent of the RDA—the typical cutoff point that defines nutrient adequacy—for energy, vitamin E, vitamin B6, calcium, magnesium and zinc. None of the nutrients examined for this study fell below 78 percent of the RDA.
Lewis and coworkers found similar results [25]. The authors examined nutrient intakes and body weights of young children (4 to 10 years of age) and adolescents (11 to 18 years of age) characterized as either high or moderate consumers of added sugars. The associations of consumption of added sugars were analyzed as grams of added sugars consumed per kilogram of body weight (g/kg) and as percentage of daily energy intake contributed by added sugars (% kcal). The high-added sugars consumers in the younger age group consumed 95 to 100 percent of the RDA for calcium.
Others have found that diets of 10-year-old children characterized as "high in total sugars" were more likely to meet current dietary recommendations for fat, saturated fat and cholesterol and were adequate with regard to most vitamins and minerals [13]. Interestingly, the children with the highest intake of total sugar consumed significantly greater amounts of beverages, milk and candy and had higher intake of calcium than did children in the lowest quartile of sugars intake.
There are several limitations to the research reported in our study. As with all cross-sectional studies, one should not infer causation based solely on the statistical associations presented here. There is also likely to be measurement error in the data set. The CSFII uses a 24-hour dietary recall method to collect data. USDA uses a computer-assisted, multiple-pass collection technique to minimize misreporting or underreporting of foods, but some measurement error is unavoidable when using self-reported dietary assessment methods. Finally, the relationship between added sugars and grains is likely to be inflated in our models because individuals who consume a large number of grain servings will have inflated estimates of added sugars consumption.
|
CONCLUSIONS |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSCONCLUSIONSACKNOWLEDGMENTSREFERENCES |
|---|
Our results for dairy and calcium consumption, in particular, cast serious doubt on the "displacement theory," that added sugars are displacing servings of dairy foods from the diets of children and adolescents or compromising their calcium intake. Of the four coefficients representing the associations of added sugars with dairy and calcium consumption for children and adolescents, only one is statistically significant. The one statistically significant coefficient shows that a one-gram increase in added sugars reduces the predicted servings of dairy by 0.0017. That means it would require an increase of more than 588 grams of added sugars to reduce a child’s predicted servings of dairy by one. The association with added sugars is so small that it would require huge behavioral changes for added sugars to have any practical association with servings of dairy.
Our model also demonstrates the importance of examining the entire diet of an individual, rather than a single ingredient like added sugars. An individual’s diet quality is affected by the totality of her or his food choices, not simply by one or two foods or ingredients of foods. Our model shows that other energy sources frequently have a stronger association with an individual’s diet quality than do added sugars.
Carbohydrates minus added sugars have a consistently positive, statistically significant and relatively larger association than added sugars with vitamin and mineral intake. Protein and fat have mixed associations that are usually statistically significant, but the size and direction of the associations vary depending on which vitamin or mineral is examined.
Alcohol has a consistently negative, statistically significant association with consumption of the food groups, vitamins and minerals examined in this paper. Moderate alcohol consumption is unlikely to have any serious impact on an individual’s diet, but heavy alcohol consumption tends to be associated with a poor diet.
Our work shows that nutritionists and policy advisors should not focus on added sugars as a unique detriment to the overall diet quality of Americans. Rather, they should consider the quality of the total diet. While it is true that added sugars are associated with certain micronutrients, the direction of the association is inconsistent. In addition, there is no association between added sugars and calcium intake among children and adolescents. This study casts serious doubt on the validity of the displacement theory as it applies to added sugars.
|
ACKNOWLEDGMENTS |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSCONCLUSIONSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSCONCLUSIONSACKNOWLEDGMENTSREFERENCES |
|---|
Received August 10, 2000. Accepted November 5, 2000.
|
REFERENCES |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSCONCLUSIONSACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Ruottinen, H. Niinikoski, H. Lagstrom, T. Ronnemaa, M. Hakanen, J. Viikari, E. Jokinen, and O. Simell High Sucrose Intake Is Associated With Poor Quality of Diet and Growth Between 13 Months and 9 Years of Age: The Special Turku Coronary Risk Factor Intervention Project Pediatrics, June 1, 2008; 121(6): e1676 - e1685. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. F. Krebs, J. H. Himes, D. Jacobson, T. A. Nicklas, P. Guilday, and D. Styne Assessment of Child and Adolescent Overweight and Obesity Pediatrics, December 1, 2007; 120(Supplement_4): S193 - S228. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. S Malik, M. B Schulze, and F. B Hu Intake of sugar-sweetened beverages and weight gain: a systematic review. Am. J. Clinical Nutrition, August 1, 2006; 84(2): 274 - 288. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. A. Forshee and M. L. Storey Controversy and Statistical Issues in the Use of Nutrient Densities in Assessing Diet Quality J. Nutr., October 1, 2004; 134(10): 2733 - 2737. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. O Fisher, D. C Mitchell, H. Smiciklas-Wright, M. L Mannino, and L. L Birch Meeting calcium recommendations during middle childhood reflects mother-daughter beverage choices and predicts bone mineral status Am. J. Clinical Nutrition, April 1, 2004; 79(4): 698 - 706. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. P Murphy and R. K Johnson The scientific basis of recent US guidance on sugars intake Am. J. Clinical Nutrition, October 1, 2003; 78(4): 827S - 833. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. M. Reynolds and M. S. Finke The Influence of Sweetened Drink Consumption on the Likelihood of Meeting the Recommended Dietary Allowance for Vitamins and Minerals Family and Consumer Sciences Research Journal, December 1, 2002; 31(2): 195 - 205. [Abstract] [PDF]
|
|
|
|
|
|
U. Alexy, W. Sichert-Hellert, and M. Kersting Fortification Masks Nutrient Dilution due to Added Sugars in the Diet of Children and Adolescents J. Nutr., September 1, 2002; 132(9): 2785 - 2791. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Franz and J. P. Bantle Response to the Diabetes Nutrition Study Group of the European Association for the Study of Diabetes Diabetes Care, July 1, 2002; 25(7): 1258 - 1259. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
indicates that a coefficient is statistically significant at less than the 0.05 level. a. The relationship between added sugars and the percentage of the Recommended Dietary Allowances for selected vitamins and minerals among individuals over the age of two. b. The relationship between added sugars and servings of major food groups among individuals over the age of two. c. The relationship between added sugars and the percentage of the Recommended Dietary Allowances for selected vitamins and minerals among children 6 to 11 years of age. d. The relationship between added sugars and servings of major food groups among children 6 to 11 years of age. e. The relationship between added sugars and the percentage of the Recommended Dietary Allowances for selected vitamins and minerals among adolescents 12 to 19 years of age. f. The relationship between added sugars and servings of major food groups among adolescents 12 to 19 years of age.